Pith. sign in

REVIEW 4 major objections 5 minor 61 references

Co-designing active amplification, element mobility, and semantic code length sharply improves vehicular semantic spectral efficiency.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-01 11:18 UTC pith:4RR2AJTE

load-bearing objection A solid engineering paper with a genuinely new architecture, but the 9.2% mobility-gain claim rests on a flawed selection step in Algorithm 1, and the monotonicity proof has a real gap. the 4 major comments →

arxiv 2607.26658 v1 pith:4RR2AJTE submitted 2026-07-29 cs.NI eess.SP

Active Movable-Element RIS Assisted Vehicular Semantic Communications: Modeling and Optimization

Maoxin Ji , Qiong Wu , Jingbo Zhang , Pingyi Fan , Kezhi Wang , Wen Chen , Guoqiang Mao , Khaled B. Letaief This is my paper
classification cs.NI eess.SP
keywords semantic communicationactive RISmovable elementsposition optimizationvehicular networkssemantic spectral efficiencyalternating optimizationreconfigurable intelligent surface
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper argues that the two standard ways of strengthening an intelligent reflecting surface for vehicular links—active amplification and movable elements—fail individually, but succeed when co-designed with the semantic coding length. It proposes a row-movable active RIS (RM-A-RIS) in which each row of reflecting elements slides along a rail and each element amplifies its reflected signal, and it jointly optimizes element positions, active reflection coefficients, and the number of symbols used per semantic message to maximize semantic spectral efficiency (SSE). In simulation the full joint design reaches 26.74 semantic units per second per hertz (suts/Hz), a 132.9% gain over a passive RIS, 9.2% over an active RIS with fixed positions, and 35.2% over a particle-swarm heuristic baseline. The paper's core argument is that active amplification raises average SINR to conquer long-distance path loss while element mobility lets the array physically escape localized deep fades; the two together provide the flat, fade-free channel that semantic decoding requires.

Core claim

The central claim is that a reconfigurable intelligent surface whose reflecting elements can be repositioned on rails and simultaneously amplify the signal can, when jointly optimized with the number of symbols used per semantic message, materially raise the semantic spectral efficiency of an uplink vehicle-to-infrastructure link. The paper models the position-dependent amplified noise, the Rician/Rayleigh fading channels, and the nonlinear lookup-table semantic similarity function, then solves the coupled non-convex problem by alternating optimization: successive convex approximation with a quadratic transform for the active reflection coefficients, a projected-gradient ascent with an order

What carries the argument

The central mechanism is the RM-A-RIS architecture and the alternating-optimization pipeline built around it. Three variables are co-optimized: the active reflection coefficient vector v (which both amplifies the signal and injects amplified thermal noise), the continuous positions U of the reflecting elements on the row-level rails (which reshape the cascaded channel and redistribute the noise), and the discrete semantic symbol length q_k per vehicle (which trades symbol count against semantic fidelity). The objective is the sum of semantic spectral efficiencies, SSE_k = (I/(q_k L)) ξ_k(γ_k, q_k), where ξ_k is the semantic similarity measured by a sentence-embedding cosine similarity, preco

Load-bearing premise

The semantic similarity lookup table is built by evaluating the semantic encoder over an AWGN channel and is then treated as a function of average SINR only; the optimization assumes this AWGN-derived mapping holds for the Rician/Rayleigh fading channels with Doppler used in the system model, and that assumption is never validated against fading-channel semantic simulations.

What would settle it

Re-run the proposed joint optimization when the semantic similarity values come from evaluating the same semantic encoder over the Rician/Rayleigh fading channels with Doppler actually used in the simulations (rather than from the AWGN lookup table); if the Sum-SSE ordering of schemes flips or the gains over passive/fixed-position active RIS drop well below the reported 132.9%/9.2%, the central claim is falsified.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

Share X Bluesky LinkedIn Reddit HN

If this is right

  • Active amplification and element mobility are complementary: the paper attributes a larger Sum-SSE gain to position optimization at similar SINR than to the SINR increase alone, suggesting that geometrically escaping fades is the main semantic benefit of mobility.
  • The two-timescale protocol makes the scheme practical: element positions are locked at frame boundaries while reflection phases and semantic lengths adapt slot-by-slot, keeping mechanical latency out of the fast-fading loop.
  • The proposed gradient-based position search outperforms discrete alternating position search and the particle-swarm heuristic, and the gap widens as the array scales to 8×8, indicating the method rather than the hardware is the bottleneck being addressed.
  • If the reported numbers hold, the joint design would make semantic communication feasible in long-distance V2I environments where both passive RIS and fixed-position active RIS fail to maintain reliable semantic decoding.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The performance gains are computed with a semantic-similarity table generated over an AWGN channel and applied to Rician/Rayleigh fading with Doppler; if the mapping from average SINR to semantic similarity is not channel-agnostic, the optimized variables solve a mismatched objective and the headline gains may shrink when the table is regenerated under fading.
  • A direct testable extension is to repopulate the lookup table by evaluating the semantic encoder over the actual fading distributions (or by online calibration) and re-running the joint optimization to see whether the ordering of schemes is preserved.
  • The semantic sensitivity weight provides a clean interface between the semantic objective and any physical-layer control, so the same alternating structure could be transplanted to other semantic tasks (image, audio, task-oriented) with a differentiable similarity measure.
  • The two-timescale design implicitly assumes vehicular geometry changes slowly relative to frame boundaries; at very high speeds or with frequent blockage, the mechanical latency may erase the spatial-diversity benefit—this crossover is not quantified in the paper.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes a row-movable active RIS (RM-A-RIS) architecture for vehicular semantic communications and formulates a joint optimization of RIS element positions, active reflection coefficients, and semantic symbol lengths to maximize Sum-SSE. The optimization is attacked with an alternating optimization framework: a semantic-sensitivity-weighted SCA/quadratic-transform phase update with Lagrange multiplier and amplitude clipping, a penalty-based projected gradient ascent with a PAVA projection for positions, and a one-dimensional greedy search for semantic symbol lengths. Simulations report Sum-SSE of 26.74 suts/Hz, with 132.9%, 9.2%, and 35.2% improvements over passive RIS, fixed-position active RIS, and QPSO benchmarks. The algebraic reformulations in Section III and the PAVA projection are largely internally consistent, and the paper provides a source-code link. However, several load-bearing gaps in the algorithm's feasibility and convergence claims, and in the channel transferability of the semantic similarity table, currently undercut the central quantitative conclusions.

Significance. If the results hold, the paper would be a useful contribution to active/movable RIS design for semantic communications: it co-designs physical-layer geometry and semantic parameters, gives a closed-form per-element phase update with low complexity, and introduces a two-timescale protocol respecting mechanical constraints. The PAVA-based exact projection onto ordered box constraints is a clean technical device, and the complexity analysis is explicit. The claimed gains are large enough to matter, but they depend on three unverified assumptions: that the position-selection step in Algorithm 1 produces feasible and monotone improvements, that the AWGN-trained semantic similarity table is valid for the Rician/Rayleigh fading channels simulated, and that the semantic similarity constraint is actually enforced. These are not cosmetic issues; they affect the validity of the headline 9.2% mobility gain and the reported Sum-SSE values.

major comments (4)
  1. [IV-D, Algorithm 1 (lines 7-16), Eq. (64)] The monotonicity proof of Theorem 1 is not established. Step 1 selects the best position U^(i+1) by evaluating the unpenalized objective f(v^(i), U^(i,iter), q^(i)) and never checks the total-power constraint (51d)/(20c) or the per-element saturation constraint (51e). The reflection vector v^(i) was feasible only for the old positions U^(i); at a candidate position it may violate the power budget, so the f-value used for selection can be unattainable. After Step 2 re-optimizes v under constraints (31b)-(31c), the constrained Sum-SSE at U^(i+1) can be lower than the value used for selection. Thus inequality (64), O(v^(t+1),U^(t+1),q) >= O(v^(t+1),U^(t),q), does not follow; the selection only guarantees O(v^(t),U^(t+1),q) >= O(v^(t),U^(t),q) for the unpenalized objective. No convergence curve of the constrained objective and no post-hoc feasibility audit of the final U* are given. The 9.2%
  2. [II-D and II-B/E, Eq. (17)] The semantic similarity lookup table xi_k(gamma_k,q_k) is generated by evaluating a pre-trained DeepSC model over an AWGN channel, but the system model uses Rician vehicle-RIS links, Rayleigh direct links, and Doppler-induced channel aging. The optimization objective (20a) assumes that the mapping from average/instantaneous SINR to BERT-level semantic similarity is channel-agnostic. No validation is provided that the AWGN table remains accurate under Rician/Rayleigh fading with time-varying SINR. If fading changes this mapping, the optimized positions, phases, and q values solve the wrong objective and the reported SSE gains are not meaningful for the claimed scenario. Please validate the lookup table by simulating DeepSC over the actual fading channels, or provide a formal/empirical justification for replacing the fading channel with an effective AWGN SINR.
  3. [III-B and IV-B, Eqs. (60)-(62)] The semantic similarity constraint (20b), xi_k >= xi_th, is not guaranteed to hold. It is explicitly relaxed during phase optimization, and in the q-subproblem the algorithm sets q_k to Q_max when the feasible set K_feasible is empty, which can still violate (20b). No final check or penalty for (20b) is reported. Since the problem is formulated as a constrained maximization, the Sum-SSE values in Figs. 4-10 may include operating points that are infeasible with respect to a named constraint, making comparisons with benchmarks invalid. Please strictly enforce (20b), or report the fraction of users/links that violate it and quantify the impact on the headline numbers.
  4. [III-A and IV-D, Eq. (21) and Theorem 1] The SCA/MM argument for the phase step is not fully supported. Equation (21) replaces SSE_k(gamma_k) by a first-order Taylor expansion at the current gamma_k^(t) and treats this as a surrogate lower bound. This is valid only if SSE_k is concave in gamma_k (or if the linearization is otherwise a global underestimator) over the relevant range. No concavity claim is stated or proved for the spline-interpolated empirical similarity function. Without a rigorous lower-bound surrogate, inequality (63) is not established. Please either prove the required concavity/majorization property or replace the analytical convergence claim with a numerical demonstration that the exact constrained objective is non-decreasing in the implemented iterations.
minor comments (5)
  1. [V, Fig. 2] The learning rate eta=0.0005 is selected on the same scenario used for the final performance results. This introduces a form of hyperparameter tuning on the test setting; please use a separate validation scenario or cross-validation, and report sensitivity with error bars.
  2. [Abstract] The phrase 'achieving up to 132.9%, 9.2%, and 35.2% improvements' is imprecise: these are point estimates at one default parameter set, not maxima over the operating range. Please state them as default-scenario improvements.
  3. [IV-E, Table I] The execution-time discussion states that a 1.384 s full joint optimization corresponds to about 27.6 m displacement at 'highway speeds', but the simulation velocity is 20 m/s (72 km/h). At highway speeds the displacement would be larger, and 27.6 m may already exceed the spatial correlation distance assumed for the fast-varying channel. Please reconcile the timescale argument with the simulation parameters.
  4. [V] All results are averaged over only 3 random seeds and 50 frames, and no confidence intervals are shown. Adding error bars or box plots would substantially strengthen the quantitative claims, especially for the small 9.2% mobility gain over Phase Only.
  5. [Throughout] Minor typographical/formatting issues include the nonstandard spacing in 'PA V A' and the notation U^(i,iter) in Algorithm 1, which is used but not formally defined. Please clean these up.

Circularity Check

0 steps flagged

No circular derivation: the SSE objective uses an external DeepSC lookup table and external benchmarks; self-citations are background only.

full rationale

The derivation chain is self-contained. The objective (20a) maximizes Sum-SSE (17), where the semantic similarity ξ_k(γ_k,q_k) is generated externally by evaluating a pre-trained DeepSC model over an AWGN channel (Section II-D) and interpolated by cubic splines; no parameter appearing in the claimed result is fitted to the reported Sum-SSE and then re-predicted. The benchmarks (passive RIS, fixed-position active RIS, QPSO, APS, Random) are external or standard baselines, and the reported gains are computed by evaluating the same objective, not by invoking a self-citation. The many self-citations in the reference list support background statements (e.g., prior vehicular/semantic/RIS work) and are not load-bearing for the central AO algorithm, Theorem 1, or the simulation conclusions. The closest concern — Algorithm 1's best-so-far position selection evaluates the unpenalized Sum-SSE f(v^(i), U^(i,iter), q^(i)) with reflection coefficients from the previous AO iteration before re-optimizing v — is a potential optimization/proof gap in the monotonicity claim (64), but it is not a circular reduction: the selected candidate is not an input to the definition of SSE, nor is any fitted parameter renamed as a prediction. Similarly, choosing η=0.0005 on the same scenario (Fig. 2) is a hyperparameter selection issue rather than circularity. Overall, no equation reduces to its own inputs, and no load-bearing result is imported from the authors' prior work.

Axiom & Free-Parameter Ledger

4 free parameters · 4 axioms · 1 invented entities

The paper's contributions depend on a set of modeling choices and algorithm hyperparameters. The most significant unvalidated axiom is the transferability of an AWGN-generated semantic similarity table to fading channels; this directly feeds the objective function. The optimization also relies on algorithmic hyperparameters (η, ρ, iteration counts) that are chosen by hand or tuned on the same scenario. No invented physical entities beyond the RM-A-RIS hardware concept.

free parameters (4)
  • Learning rate η = 0.0005
    Selected as peak Sum-SSE in Fig. 2 on the same simulation scenario used for all reported results; a form of tuning on the target setting.
  • Penalty factor ρ = not specified
    Controls enforcement of the total power constraint in the penalty method; value not given, and no feasibility study is reported, making the final solution's constraint satisfaction unverifiable.
  • AO/hyper-parameters (I_AO, I_New, T_pos, learning schedule) = I_AO=3, I_New=5, T_pos=20
    Truncation/hyperparameters chosen by hand; convergence properties depend on them.
  • Unspecified channel/hardware parameters (K_sr, σ_sh, F_RIS, bandwidth, carrier frequency) = not listed in Table II
    Required to reproduce the simulations; absence weakens reproducibility.
axioms (4)
  • domain assumption The semantic similarity lookup table created under AWGN is a valid proxy for semantic similarity in Rician/Rayleigh fading vehicular channels.
    Used to define SSE in Eq. (17) and to optimize/evaluate all schemes; no validation is provided.
  • domain assumption The RIS-to-BS link is deterministic LoS and perfectly known once U is fixed; the direct and vehicle-RIS links follow the specified Rician/Rayleigh models with Jakes channel aging.
    These channel models define the simulation; results are conditional on them.
  • domain assumption Active RIS amplifiers are independent noise sources with given P_sat, A_max, and total power budget; hardware can mechanically move rows within frame boundaries.
    Hardware feasibility assumed, not demonstrated.
  • standard math Standard optimization results: SCA surrogate is a valid minorizer, quadratic transform is exact for sum-of-ratios, PAVA gives an exact Euclidean projection onto ordered boxes, and a bounded monotone sequence converges.
    Used in Sections III and IV; standard and generally accepted.
invented entities (1)
  • RM-A-RIS (Row-Movable Active RIS) no independent evidence
    purpose: Hardware architecture combining active reflection amplification with row-level element mobility to enhance spatial diversity and SINR for vehicular semantic links.
    Only simulated; no prototype, channel measurement, or mechanical-latency experiment is provided, so its practical feasibility and gains are unverified.

pith-pipeline@v1.3.0-daily-deepseek · 24675 in / 14474 out tokens · 137147 ms · 2026-08-01T11:18:56.804609+00:00 · methodology

0 comments
Cite this review

Pith. "Pith review of Active Movable-Element RIS Assisted Vehicular Semantic Communications: Modeling and Optimization." pith.science (2026). https://pith.science/paper/4RR2AJTE

@misc{pith2026260726658,
  author       = {Pith},
  title        = {Pith review of: Active Movable-Element RIS Assisted Vehicular Semantic Communications: Modeling and Optimization},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4RR2AJTE}},
  note         = {Machine review of arXiv:2607.26658}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Severe signal blockage and fast-varying channels in vehicular environments pose critical challenges to reliable semantic communication. To address these, this paper proposes a novel Row-Movable Active Reconfigurable Intelligent Surface (RM-A-RIS) assisted vehicular semantic communication system. This architecture uniquely combines active signal amplification with element mobility to compensate for multiplicative fading and reconstruct channel geometry, thereby enhancing spatial diversity. We formulate a joint optimization problem to maximize Semantic Spectral Efficiency (SSE) by coordinating RIS element positions, active reflection coefficients, and semantic symbol length. An efficient Alternating Optimization (AO) algorithm is developed to tackle the coupled non-convexity. Simulation results demonstrate that the proposed scheme substantially outperforms existing benchmarks, achieving up to 132.9%, 9.2%, and 35.2% improvements in Sum-Semantic Spectral Efficiency (Sum-SSE) compared to the passive RIS, fixed-position active RIS, and QPSO baselines, respectively.

Figures

Figures reproduced from arXiv: 2607.26658 by Guoqiang Mao, Jingbo Zhang, Kezhi Wang, Khaled B. Letaief, Maoxin Ji, Pingyi Fan, Qiong Wu, Wen Chen.

Figure 1
Figure 1. Figure 1: System model and phase shifts. To realistically reflect the complex urban vehicular environment characterized by frequent blockages and scattering, we model the vehicle-to-BS direct links as Rayleigh fading, the vehicle-to-RIS links as Rician fading, and the elevated static RIS-to-BS links as Line-of-Sight (LoS) dominated channels. The specific channel matrices are mod￾eled as follows. 1) Vehicle-to-RIS Ch… view at source ↗
Figure 2
Figure 2. Figure 2: Learning rate sensitivity of PGA [PITH_FULL_IMAGE:figures/full_fig_p013_2.png] view at source ↗
Figure 5
Figure 5. Figure 5: Average Sum-SSE for different number of vehicles [PITH_FULL_IMAGE:figures/full_fig_p014_5.png] view at source ↗
Figure 8
Figure 8. Figure 8: Impact of CSI estimation errors on the Sum-SSE of different optimization schemes [PITH_FULL_IMAGE:figures/full_fig_p014_8.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

61 extracted references

  1. [1]

    V2X-assisted distributed computing and control framework for connected and automated CA Vs under ramp merging scenario,

    J. Chu, Q. Wu, P. Fan, W. Chen, K. Wang, N. Cheng, and K. B. Letaief, “V2X-assisted distributed computing and control framework for connected and automated CA Vs under ramp merging scenario,”IEEE Trans. Mobile Comput., vol. 25, no. 5, pp. 7436–7453, 2026

    2026

  2. [2]

    Toward 6G networks: Use cases and technologies,

    M. Giordani, M. Polese, M. Mezzavilla, S. Rangan, and M. Zorzi, “Toward 6G networks: Use cases and technologies,”IEEE Commun. Mag., vol. 58, no. 3, pp. 55–61, 2020

    2020

  3. [3]

    Large language model-based task offloading and resource allocation for digital twin edge computing networks,

    Q. Wu, Y . Xie, P. Fan, D. Qin, K. Wang, N. Cheng, and K. B. Letaief, “Large language model-based task offloading and resource allocation for digital twin edge computing networks,”IEEE Trans. Mobile Comput., early access, 2026, doi: 10.1109/TMC.2026.3664866

    arXiv 2026

  4. [4]

    RadioDiff-k2: Helmholtz equation informed generative diffusion model for multi-path aware radio map construction,

    X. Wang, Q. Zhang, N. Cheng, R. Sun, Z. Li, S. Cui, and X. Shen, “RadioDiff-k2: Helmholtz equation informed generative diffusion model for multi-path aware radio map construction,”IEEE J. Sel. Areas Commun., early access, 2025. [Online]. Available: https://doi.org/10. 1109/JSAC.2025.3641105

    arXiv 2025

  5. [5]

    Trajectory pro- tection schemes based on a gravity mobility model in IoT,

    Q. Wu, H. Liu, C. Zhang, Q. Fan, Z. Li, and K. Wang, “Trajectory pro- tection schemes based on a gravity mobility model in IoT,”Electronics, vol. 8, no. 2, p. 148, 2019

    2019

  6. [6]

    Performance modeling and analysis of the AD- HOC MAC protocol for V ANETs,

    Q. Wu and J. Zheng, “Performance modeling and analysis of the AD- HOC MAC protocol for V ANETs,” inProc. IEEE Int. Conf. Commun. (ICC), London, UK, 2015, pp. 3646–3652

    2015

  7. [7]

    Performance analysis of IEEE 802.11 p for continuous backoff freezing in IoV ,

    Q. Wu, S. Xia, Q. Fan, and Z. Li, “Performance analysis of IEEE 802.11 p for continuous backoff freezing in IoV ,”Electronics, vol. 8, no. 12, p. 1404, 2019

    2019

  8. [8]

    A swarming approach to optimize the one-hop delay in smart driving inter-platoon communications,

    Q. Wu, S. Nie, P. Fan, H. Liu, F. Qiang, and Z. Li, “A swarming approach to optimize the one-hop delay in smart driving inter-platoon communications,”Sensors, vol. 18, no. 10, p. 3307, 2018

    2018

  9. [9]

    Performance modeling and analysis of the ADHOC MAC protocol for vehicular networks,

    Q. Wu and J. Zheng, “Performance modeling and analysis of the ADHOC MAC protocol for vehicular networks,”Wireless Netw., vol. 22, no. 3, pp. 799–812, 2016

    2016

  10. [10]

    Performance modeling and analysis of IEEE 802.11 DCF based fair channel access for vehicle-to-roadside commu- nication in a non-saturated state,

    Q. Wu and J. Zheng, “Performance modeling and analysis of IEEE 802.11 DCF based fair channel access for vehicle-to-roadside commu- nication in a non-saturated state,”Wireless Netw., vol. 21, no. 1, pp. 1–11, 2015

    2015

  11. [11]

    Performance modeling of the IEEE 802.11 p EDCA mechanism for V ANET,

    Q. Wu and J. Zheng, “Performance modeling of the IEEE 802.11 p EDCA mechanism for V ANET,” inProc. IEEE Global Commun. Conf. (GLOBECOM), Austin, TX, USA, 2014, pp. 57–63

    2014

  12. [12]

    DRL- based optimization for AoI and energy consumption in C-V2X enabled IoV ,

    Z. Zhang, Q. Wu, P. Fan, N. Cheng, W. Chen, and K. B. Letaief, “DRL- based optimization for AoI and energy consumption in C-V2X enabled IoV ,”IEEE Trans. Green Commun. Netw., early access, 2025

    2025

  13. [13]

    Study on refined deployment of wireless mesh sensor network,

    J. Fan, S. Yin, Q. Wu, and F. Gao, “Study on refined deployment of wireless mesh sensor network,” inProc. 6th Int. Conf. Wireless Commun. Netw. Mobile Comput. (WiCOM), Chengdu, China, 2010, pp. 1–5

    2010

  14. [14]

    Optimal cooperative beamforming design for MIMO decode-and-forward relay channels,

    K. Xiong, P. Fan, Z. Xu, H. C. Yang, and K. B. Letaief, “Optimal cooperative beamforming design for MIMO decode-and-forward relay channels,”IEEE Trans. Signal Process., vol. 62, no. 6, pp. 1476–1489, 2014

    2014

  15. [15]

    Doppler frequency offset estimation and diversity reception scheme of high-speed railway with multiple antennas on separated carriage,

    Y . Yang and P. Fan, “Doppler frequency offset estimation and diversity reception scheme of high-speed railway with multiple antennas on separated carriage,”J. Mod. Transport., vol. 20, no. 4, pp. 227–233, 2012

    2012

  16. [16]

    Global proportional fair scheduling for networks with multiple base stations,

    H. Zhou, P. Fan, and J. Li, “Global proportional fair scheduling for networks with multiple base stations,”IEEE Trans. Veh. Technol., vol. 60, no. 4, pp. 1867–1879, 2011

    2011

  17. [17]

    Delay-constrained optimal link scheduling in wireless sensor networks,

    Q. Wang, D. O. Wu, and P. Fan, “Delay-constrained optimal link scheduling in wireless sensor networks,”IEEE Trans. Veh. Technol., vol. 59, no. 9, pp. 4564–4577, 2010

    2010

  18. [18]

    Network coding for two-way relaying networks over Rayleigh fading channels,

    W. Li, J. Li, and P. Fan, “Network coding for two-way relaying networks over Rayleigh fading channels,”IEEE Trans. Veh. Technol., vol. 59, no. 9, pp. 4476–4488, 2010

    2010

  19. [19]

    Network coding for efficient multicast routing in wireless ad-hoc networks,

    J. Zhang, P. Fan, and K. B. Letaief, “Network coding for efficient multicast routing in wireless ad-hoc networks,”IEEE Trans. Commun., vol. 56, no. 4, pp. 598–607, 2008

    2008

  20. [20]

    A neighbor-table-based multipath routing in ad hoc networks,

    Z. Yao, J. Jiang, P. Fan, Z. Cao, and V . O. K. Li, “A neighbor-table-based multipath routing in ad hoc networks,” inProc. 57th IEEE Semiannu. Veh. Technol. Conf. (VTC Spring), Jeju, South Korea, 2003, pp. 1739– 1743

    2003

  21. [21]

    Investigation of the time-offset- based QoS support with optical burst switching in WDM networks,

    P. Fan, C. Feng, Y . Wang, and N. Ge, “Investigation of the time-offset- based QoS support with optical burst switching in WDM networks,” in Proc. IEEE Int. Conf. Commun. (ICC), New York, NY , USA, 2002, pp. 2682–2686

    2002

  22. [22]

    Block coded modulation for the reduction of the peak to average power ratio in OFDM systems,

    P. Fan and X.-G. Xia, “Block coded modulation for the reduction of the peak to average power ratio in OFDM systems,” inProc. IEEE Wireless Commun. Netw. Conf. (WCNC), New Orleans, LA, USA, 1999, pp. 1095–1099

    1999

  23. [23]

    High stable and accurate vehicle selection scheme based on federated edge learning in vehicular networks,

    Q. Wu, X. Wang, Q. Fan, P. Fan, C. Zhang, and Z. Li, “High stable and accurate vehicle selection scheme based on federated edge learning in vehicular networks,”China Commun., vol. 20, no. 3, pp. 1–17, 2023

    2023

  24. [24]

    Optimal resource allocation in wireless powered communication networks with 16 user cooperation,

    X. Di, K. Xiong, P. Fan, H. C. Yang, and K. B. Letaief, “Optimal resource allocation in wireless powered communication networks with 16 user cooperation,”IEEE Trans. Wireless Commun., vol. 16, no. 12, pp. 7936–7949, 2017

    2017

  25. [25]

    Deep learning-based semantic communication system for wireless image transmission,

    L. Hu, L. Yu, and Z. Qin, “Deep learning-based semantic communication system for wireless image transmission,”IEEE Wireless Commun. Lett., vol. 14, no. 8, pp. 2391–2395, 2025

    2025

  26. [26]

    Semantic communication for the Internet of Vehicles: A multiuser cooperative approach,

    W. Xu, Y . Zhang, F. Wang, Z. Qin, C. Liu, and P. Zhang, “Semantic communication for the Internet of Vehicles: A multiuser cooperative approach,”IEEE Veh. Technol. Mag., vol. 18, no. 1, pp. 100–109, 2023

    2023

  27. [27]

    Semantic-aware spectrum sharing in Internet of Vehicles based on deep reinforcement learning,

    Z. Shao, Q. Wu, P. Fan, N. Cheng, W. Chen, J. Wang, and K. B. Letaief, “Semantic-aware spectrum sharing in Internet of Vehicles based on deep reinforcement learning,”IEEE Internet Things J., vol. 11, no. 23, pp. 38 521–38 536, 2024

    2024

  28. [28]

    Multi-UA V-assisted MEC in Internet of Vehicles with combined multi-modal semantic communi- cation under jamming attacks,

    S. Liu, H. Yang, M. Zheng, and L. Xiao, “Multi-UA V-assisted MEC in Internet of Vehicles with combined multi-modal semantic communi- cation under jamming attacks,”IEEE Trans. Mobile Comput., vol. 24, no. 8, pp. 7600–7614, 2025

    2025

  29. [29]

    Emotion-aided semantic communication system for reliable semantic recovery under low SNR,

    P. Luo, H. Zhao, K. Cao, Y . Liu, Y . Zhang, and J. Wei, “Emotion-aided semantic communication system for reliable semantic recovery under low SNR,”IEEE Commun. Lett., vol. 28, no. 3, pp. 503–507, 2024

    2024

  30. [30]

    Velocity-adaptive access scheme for semantic-aware vehicular networks: Joint fairness and AoI optimization,

    X. Xu, Q. Wu, P. Fan, K. Wang, N. Cheng, W. Chen, and K. B. Letaief, “Velocity-adaptive access scheme for semantic-aware vehicular networks: Joint fairness and AoI optimization,”IEEE Trans. Mobile Comput., early access, 2026, doi: 10.1109/TMC.2026.3667698

    arXiv 2026

  31. [31]

    Semantic-aware cooperative communication and computation framework in vehicular networks,

    J. Zhang, M. Ji, Q. Wu, P. Fan, K. Wang, and W. Chen, “Semantic-aware cooperative communication and computation framework in vehicular networks,”IEEE Netw. Lett., early access, 2025. [Online]. Available: https://doi.org/10.1109/LNET.2025.3648419

    arXiv 2025

  32. [32]

    Resource allocation for twin maintenance and task processing in vehicular edge computing network,

    Y . Xie, Q. Wu, P. Fan, N. Cheng, W. Chen, J. Wang, and K. B. Letaief, “Resource allocation for twin maintenance and task processing in vehicular edge computing network,”IEEE Internet Things J., vol. 12, no. 15, pp. 32008–32021, 2025

    2025

  33. [33]

    DRL-based resource allocation for motion blur resistant federated self- supervised learning in IoV ,

    X. Gu, Q. Wu, P. Fan, Q. Fan, N. Cheng, W. Chen, and K. B. Letaief, “DRL-based resource allocation for motion blur resistant federated self- supervised learning in IoV ,”IEEE Internet Things J., vol. 12, no. 6, pp. 7067–7085, 2025

    2025

  34. [34]

    CDDM: Channel denoising diffusion models for wireless semantic communications,

    T. Wu, Z. Chen, D. He, L. Qian, Y . Xu, M. Tao, and W. Zhang, “CDDM: Channel denoising diffusion models for wireless semantic communications,”IEEE Trans. Wireless Commun., vol. 23, no. 9, pp. 11 168–11 183, 2024

    2024

  35. [35]

    A generalized semantic communication system: From sources to channels,

    Z. Qin, F. Gao, B. Lin, X. Tao, G. Liu, and C. Pan, “A generalized semantic communication system: From sources to channels,”IEEE Wireless Commun., vol. 30, no. 3, pp. 18–26, 2023

    2023

  36. [36]

    Towards smart and reconfigurable environment: Intelligent reflecting surface aided wireless network,

    Q. Wu and R. Zhang, “Towards smart and reconfigurable environment: Intelligent reflecting surface aided wireless network,”IEEE Commun. Mag., vol. 58, no. 1, pp. 106–112, 2020

    2020

  37. [37]

    Smart radio environments empowered by reconfigurable intelligent surfaces: How it works, state of research, and the road ahead,

    M. Di Renzo, A. Zappone, M. Debbah, M.-S. Alouini, C. Yuen, J. De Rosny, and S. Tretyakov, “Smart radio environments empowered by reconfigurable intelligent surfaces: How it works, state of research, and the road ahead,”IEEE J. Sel. Areas Commun., vol. 38, no. 11, pp. 2450–2525, 2020

    2020

  38. [38]

    Reconfigurable intelligent surfaces for energy efficiency in wireless communication,

    C. Huang, A. Zappone, G. C. Alexandropoulos, M. Debbah, and C. Yuen, “Reconfigurable intelligent surfaces for energy efficiency in wireless communication,”IEEE Trans. Wireless Commun., vol. 18, no. 8, pp. 4157–4170, 2019

    2019

  39. [39]

    Multi-agent DRL-based task offloading in multiple RIS-aided IoV networks,

    B. Hazarika, K. Singh, S. Biswas, S. Mumtaz, and C.-P. Li, “Multi-agent DRL-based task offloading in multiple RIS-aided IoV networks,”IEEE Trans. Veh. Technol., vol. 73, no. 1, pp. 1175–1190, 2023

    2023

  40. [40]

    Deep-reinforcement-learning-based AoI-aware resource allocation for RIS-aided IoV networks,

    K. Qi, Q. Wu, P. Fan, N. Cheng, W. Chen, J. Wang, and K. B. Letaief, “Deep-reinforcement-learning-based AoI-aware resource allocation for RIS-aided IoV networks,”IEEE Trans. Veh. Technol., vol. 74, no. 1, pp. 1365–1378, 2025

    2025

  41. [41]

    Active reconfigurable intelligent surface-aided wireless communications,

    R. Long, Y .-C. Liang, Y . Pei, and E. G. Larsson, “Active reconfigurable intelligent surface-aided wireless communications,”IEEE Trans. Wire- less Commun., vol. 20, no. 8, pp. 4962–4975, 2021

    2021

  42. [42]

    Active RIS versus passive RIS: Which is superior with the same power budget?

    K. Zhi, C. Pan, H. Ren, K. K. Chai, and M. Elkashlan, “Active RIS versus passive RIS: Which is superior with the same power budget?” IEEE Commun. Lett., vol. 26, no. 5, pp. 1150–1154, 2022

    2022

  43. [43]

    Active RIS vs. passive RIS: Which will prevail in 6G?

    Z. Zhang, L. Dai, X. Chen, C. Liu, F. Yang, R. Schober, and H. V . Poor, “Active RIS vs. passive RIS: Which will prevail in 6G?”IEEE Trans. Commun., vol. 71, no. 3, pp. 1707–1725, 2023

    2023

  44. [44]

    Active RIS-aided massive MIMO with im- perfect CSI and phase noise,

    Z. Peng, J. Zhu, C. Pan, Z. Zhang, D. Benevides da Costa, M. Elkashlan, and G. K. Karagiannidis, “Active RIS-aided massive MIMO with im- perfect CSI and phase noise,”IEEE Trans. Wireless Commun., vol. 24, no. 8, pp. 6824–6840, 2025

    2025

  45. [45]

    RIS-assisted green secure communications: Active RIS or passive RIS?

    W. Lv, J. Bai, Q. Yan, and H. M. Wang, “RIS-assisted green secure communications: Active RIS or passive RIS?”IEEE Wireless Commun. Lett., vol. 12, no. 2, pp. 237–241, 2023

    2023

  46. [46]

    Beamforming saturation in two-timescale RIS-assisted communication,

    M. Sadeghian, A. Lozano, and G. Fodor, “Beamforming saturation in two-timescale RIS-assisted communication,” inProc. IEEE Int. Work- shop Signal Process. Adv. Wireless Commun. (SPAWC), Surrey, U.K., 2025, pp. 1–5

    2025

  47. [47]

    RIS-aided beamforming design for MIMO systems via unified manifold optimization,

    K. Zhong, J. Hu, H. Li, Y . Zhou, and X. Chen, “RIS-aided beamforming design for MIMO systems via unified manifold optimization,”IEEE Trans. Veh. Technol., vol. 74, no. 1, pp. 674–685, 2025

    2025

  48. [48]

    Fluid antenna systems,

    K.-K. Wong, A. Shojaeifard, K.-F. Tong, and Y . Zhang, “Fluid antenna systems,”IEEE Trans. Wireless Commun., vol. 20, no. 3, pp. 1950–1962, 2021

    1950

  49. [49]

    STAR-RIS- empowered integrated sensing and covert communication system with movable elements: Joint robust beamforming and element deployment design,

    T. Zhou, K. Xu, G. Hu, C. Li, X. Xia, C. Wei, and Y . Chen, “STAR-RIS- empowered integrated sensing and covert communication system with movable elements: Joint robust beamforming and element deployment design,”IEEE Trans. Cogn. Commun. Netw., vol. 11, no. 5, pp. 2893– 2909, 2025

    2025

  50. [50]

    Movable-element RIS: Joint element positioning and beamforming optimization,

    D. Zhou, W. Mei, Z. Bai, N. Li, and T. Q. Quek, “Movable-element RIS: Joint element positioning and beamforming optimization,”IEEE Wireless Commun. Lett., vol. 15, pp. 915–919, 2025

    2025

  51. [51]

    Robust and secure transmission for movable-RIS assisted ISAC with imperfect sense estimation,

    L. Zhuang, X. Xie, F. Fang, A. Attaran, and Z. Zhang, “Robust and secure transmission for movable-RIS assisted ISAC with imperfect sense estimation,”arXiv preprint arXiv:2512.20071, 2025. [Online]. Available: https://arxiv.org/abs/2512.20071

    arXiv 2025

  52. [52]

    Joint active and passive beamforming for RIS-aided semantic communication,

    Y . Huang, C. Cai, X. Yuan, and Y .-J. A. Zhang, “Joint active and passive beamforming for RIS-aided semantic communication,”IEEE Trans. Veh. Technol., vol. 73, no. 12, pp. 19815–19820, 2024

    2024

  53. [53]

    IRS-enhanced secure semantic communication networks: Cross-layer and context-awared re- source allocation,

    L. Wang, W. Wu, F. Zhou, Z. Qin, and Q. Wu, “IRS-enhanced secure semantic communication networks: Cross-layer and context-awared re- source allocation,”IEEE Trans. Wireless Commun., vol. 24, no. 1, pp. 494–508, 2025

    2025

  54. [54]

    Fluid antenna- enabled near-field integrated sensing, computing, and semantic commu- nication for emerging applications,

    Y . Yang, J. Zhou, Z. Yang, and M. R. Shikh-Bahaei, “Fluid antenna- enabled near-field integrated sensing, computing, and semantic commu- nication for emerging applications,”IEEE Trans. Cogn. Commun. Netw., vol. 11, no. 5, pp. 3062–3078, 2025

    2025

  55. [55]

    Deep learning enabled semantic communication systems,

    H. Xie, Z. Qin, G. Y . Li, and B.-H. Juang, “Deep learning enabled semantic communication systems,”IEEE Trans. Signal Process., vol. 69, pp. 2663–2675, 2021

    2021

  56. [56]

    Exploiting movable elements of intelligent reflecting surface for enhancement of integrated sensing and communication,

    X. Peng, Q. Tao, Y . L. Guan, and X. Chen, “Exploiting movable elements of intelligent reflecting surface for enhancement of integrated sensing and communication,”IEEE Trans. Wireless Commun., vol. 25, pp. 1717– 1733, 2026

    2026

  57. [57]

    Resource allocation for text semantic communications,

    L. Yan, Z. Qin, R. Zhang, Y . Li, and G. Y . Li, “Resource allocation for text semantic communications,”IEEE Wireless Commun. Lett., vol. 11, no. 7, pp. 1394–1398, 2022

    2022

  58. [58]

    DRL- based optimization for AoI and energy consumption in C-V2X enabled IoV ,

    Z. Zhang, Q. Wu, P. Fan, N. Cheng, W. Chen, and K. B. Letaief, “DRL- based optimization for AoI and energy consumption in C-V2X enabled IoV ,”IEEE Trans. Green Commun. Netw., vol. 9, no. 4, pp. 2144–2159, 2025

    2025

  59. [59]

    Enhanced velocity-adaptive scheme: Joint fair access and age of information optimization in vehicular networks,

    X. Xu, Q. Wu, P. Fan, K. Wang, N. Cheng, W. Chen, and K. B. Letaief, “Enhanced velocity-adaptive scheme: Joint fair access and age of information optimization in vehicular networks,”IEEE Trans. Mobile Comput., vol. 25, no. 3, pp. 3488–3505, 2026

    2026

  60. [60]

    Robust beamforming design for intelligent reflecting surface-aided massive MIMO system with channel aging,

    Y . He, Y . Zhu, H. Zhang, S. Xu, and Z. Xu, “Robust beamforming design for intelligent reflecting surface-aided massive MIMO system with channel aging,”IEEE Trans. Veh. Technol., early access, 2025. [Online]. Available: https://ieeexplore.ieee.org/document/11119330

    arXiv 2025

  61. [61]

    A hybrid quantum-behaved particle swarm optimization algorithm for solving inverse scattering problems,

    C. X. Yang, J. Zhang, and M. S. Tong, “A hybrid quantum-behaved particle swarm optimization algorithm for solving inverse scattering problems,”IEEE Trans. Antennas Propag., vol. 69, no. 9, pp. 5861– 5869, 2021

    2021